kimi-k2.5-thinking: benchmark scores, price and ranking

kimi-k2.5-thinking from Moonshot AI is a recent or sparsely tested model with 2 benchmark results on record; it has not been measured on enough benchmarks shared with other models to receive an aggregate rank. This page lists every score we have, based on 2 benchmark results captured up to 2026-10-13.

Organisation: Moonshot AI
Type: Open-weight / local model
Open weights: Yes
Cheapest API price: $0.450 in / $2.25 out per million tokens (via deepinfra, observed 2026-10-13T21:47Z)
Aggregate rank: Not ranked yet (too few shared benchmarks)

Across the 2 benchmarks where at least ten models are scored, kimi-k2.5-thinking is in the top 10% on 0 and in the top half on 0.

Its cheapest listed input price, $0.450 per million tokens, is lower than that of 52% of the 427 models with a listed price.

kimi-k2.5-thinking benchmark results

kimi-k2.5-thinking benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
LMArena Document Elo	1,430	#12 of 18	LMArena	2026-10-13
LMArena Vision Elo	1,252	#21 of 22	LMArena	2026-10-05

About kimi-k2.5-thinking

kimi-k2.5-thinking is an open-weight model that can run on your own hardware from Moonshot AI. The cheapest API listing we track charges $0.450 per million input tokens and $2.25 per million output tokens, through deepinfra. We hold 2 benchmark results for it across 1 area, and none is in the top tenth of its board.

Where kimi-k2.5-thinking stands

kimi-k2.5-thinking has 2 results so far and shares too few benchmarks with other models to get an aggregate rank. The rank will appear on its own once more scores are published.

kimi-k2.5-thinking, benchmark by benchmark

kimi-k2.5-thinking lands below the middle on LMArena Document Elo: 1,430, #12 of 18, with 61% of the field scoring higher. Claude Fable 5.1 (max) leads at 1,513; the closest result above is Gemini 3 Pro at 1,434. It covers crowd-voted questions about documents. The board median is 1,440, so kimi-k2.5-thinking is 10 points below it. Source: LMArena, captured 2026-10-13.

LMArena Vision Elo is a weak spot. kimi-k2.5-thinking scores 1,252, #21 of 22, behind 91% of the models measured. The leader, Gemini 3.7 Flash (high), reaches 1,296. It covers crowd-voted image understanding. The board median is 1,279, so kimi-k2.5-thinking is 27 points below it. Source: LMArena, captured 2026-10-05.

Strengths and weaknesses

General knowledge and chat: weak, ahead of 19% of the field on average across 2 benchmarks. Its best board here is LMArena Document Elo (#12 of 18) and its weakest is LMArena Vision Elo (#21 of 22).

Reading these numbers

The 2 scores on this page come from 1 source: LMArena (2). No official leaderboard supplied any of them, so treat them as third-party measurements. With 2 results the aggregate rank can move a lot when one new score arrives.